Bioinformatics: Definitions, Challenges and Impact on Health Care Systems

1
Bioinformatics Definitions, Challenges and
Impact on Health Care Systems

• Joyce Mitchell, PhD
• Professor and Chair
• Department of Biomedical Informatics
• University of Utah School of Medicine
• http//uuhsc.utah.edu/medinfo

2
Topics

• What is Bioinformatics?
• Scope of Bioinformatics
• Genomics
• Proteomics
• Functional genomics
• Genomics data and patient care
• Impact of Bioinformatics on Health Information
  Systems
• What is coming?

3
Central Dogma of Molecular Biology
Transcription
DNA
RNA
Protein
Phenotype
Phenotype
Translation
Post Translational Modification
Replication
4
What is Bioinformatics?

• Definitions

5
NIH Definition

• Bioinformatics applies principles of information
  sciences and technologies to make the vast,
  diverse, and complex life sciences data more
  understandable and useful.

6
NIH Definition cont

• Bioinformatics Research, development, or
  application of computational tools and approaches
  for expanding the use of biological, medical,
  behavioral or health data, including those to
  acquire, store, organize, archive, analyze, or
  visualize such data.
• http//www.bisti.nih.gov/CompuB
  ioDef.pdf

7
AnotherNCBI (National Center for Biotechnology
Information)

• Bioinformatics is the field of science in
  which biology, computer science, and information
  technology merge into a single discipline. The
  ultimate goal of the field is to enable the
  discovery of new biological insights and to
  create a global perspective from which unifying
  principles in biology can be discerned.
• http//www.ncbi.nlm.nih.gov/About/primer/bioinform
  atics.html

8
Bioinformatics Health Informatics

• Bioinformatics is the study of the flow of
  information in biological sciences.
• Health Informatics is the study of the flow of
  information in patient care.
• These two field are on a collision course as
  genomics data becomes used in patient care.
• Russ Altman,MD, PhD, Stanford Univ.

9
(No Transcript)
10
Scope of Bioinformatics

• OMES and OMICS

11
Omes and Omics

• Genomics
• Primarily sequences (DNA and RNA)
• Databanks and search algorithms
• Supports studies of molecular evolution (Tree
  wars)
• Proteomics
• Sequences (Protein) and structures
• Mass spectrometry, X-ray crystallography
• Databanks, knowledge bases, visualization
• Functional Genomics (transcriptomics)
• Microarray data (and SNP Chips)
• Databanks, analysis tools, controlled
  terminologies
• Genetic Epidemiology finding gene-disease
  associations
• Linkage studies
• GWAS studies (genome wide association studies)
• Systems Biology (metabolomics)
• Metabolites and interacting systems
  (interactomics)
• Graphs, visualization, modeling, networks of
  entities

12
Central Dogma of Molecular Biology
DNA
RNA
Protein
Phenotype
Phenotype
Functional Genomics (Transcriptomics)
Structural Genomics
Phenomics
Proteomics
13
Human Genome Project

• Human Genome Project - International research
  effort
• Determine sequence of human genome and other
  model organisms
• Began 1990, completed 2003
• We are now in the Post-Genomic era
• Next steps for 20,000 genes
• Function and regulation of all genes
• Significance of variations between people
• Cures, therapies, genomic healthcare

14
Genome and Genomics

• Genome entire complement of DNA in a species
• Both nuclear and mitochondrial/chloroplast
• Variants among individuals
• Genomics study of the sequence, structure and
  function of the genome. Study relationships
  among sets of genes rather than single genes.
• Comparative genomics study of the differences
  among species. Usually covers evolutionary
  studies of differences conservation over time.

15
Genome Databases (e.g., GenBank)

• Consists of
• long strings of DNA bases ATCG..
• Annotations of this database to attach meaning to
  the sequence data.
• Example entry from GenBank
• http//www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?val
  NM_000410doptgb Hemochromatosis gene HFE

16
The Genome Sequence is at handso?
The good news is that we have the human genome.
The bad news is its just a parts list
17
The Human Genome Project has catalyzed striking
paradigm changes in biology - biology is now an
information science.

• Leroy Hood, MD, PhD
• Institute for Systems Biology
• Seattle, Washington

18
Genomes In Public Databases

• Published complete genomes
• Ongoing prokaryotic genome projects
• Ongoing eukaryotic genomes

2700
http//www.genomesonline.org/
19
Genomics activities

• Sequence the genes and chromosomes done by
  breaking the DNA into parts
• Map the location of various gene entities to
  establish their order
• Compare the sequences with other known sequences
  to determine similarity
• Across species, conserved sequence motifs
• Predict secondary structure of proteins
• Create large databases GenBank, EMBL, DDBJ
• Develop algorithms and similarity measures
• BLAST and its many forms

20
Central Dogma of Molecular Biology
DNA
RNA
Protein
Phenotype
Phenotype
Genomics
Proteomics
Transcriptomics Functional Genetics
21
Proteome vs Transcriptome

• Functional genomics (transcriptomics) looks at
  the timing and regulation of gene products (mRNA,
  primarily)
• Proteome is final end-product (set of many or all
  proteins).
• Relationship between transcriptome and proteome
  is complex, due to longevity of mRNA signal,
  subsequent control of translation to protein, and
  post translational modifications.

22
Functional Genomics TechnologiesGene Chips,
Microarrays, etc
23
Functional Genomics Microarrays

• Transcriptome and transcriptomics
• High throughput technique designed to measure the
  relative abundance of mRNA in a cell or tissue
  in response to an experiment.
• Also called gene expression analysis
• (and multiple kinds of microarrays)

24
Gene Chips (SNP-Chips)

• High throughput technique designed to measure
  whether or not a sample of tissue has various
  SNPs in its DNA.
• The Gene Chip has small segments of DNA on the
  chip with known variants (SNPs). 1 million
  SNPs per chip

25
GeneChip synthesis
26
GeneChip synthesis

• Now have RNA, DNA and protein chips on the same
  piece of equipment.
• 25 DNA or RNA bases or Amino Acids in each of 1
  million spots
• DNA microarray synthesis
• http//www.hhmi.org/biointeractive/media/gene_chip
  s-lg.mov

27
Characteristics of Array Data

• Voluminous tens of thousands of variables with
  relatively few observations of each (upside down
  vs. classical biostatistics)
• Noisy error rates up to 8
• Methods designed to detect patterns and
  associations always find patterns and
  associations

28
Experimental Design

• A fundamental challenge of microarray
  experiments underdetermined systems

Kohane IS, Kho AT, Butte AJ. Microarrays for an
Integrative Genomics. (The MIT Press Cambridge,
MA 2003), p. 11.
29
GWAS Studies

• Genome Wide Association Studies
• Use SNP chips data to look for associations
  between SNP profiles and diseases.
• Beyond single gene studies into multiple gene
  studies. Usually common diseases.
• Beyond family studies into population studies
• Some analyses on combining GWAS analyses for
  diseases in pedigrees.

30
Published Genome-Wide Associations through
3/2009, 398 published GWA at p lt
5 x 10-8
NHGRI GWA Catalog www.genome.gov/GWAStudies
31
Public Microarray Data Repositories

• Major public repositories
• GEO (NCBI)
• http//www.ncbi.nlm.nih.gov/geo/
• ArrayExpress (EBI)
• http//www.ebi.ac.uk/arrayexpress/

32
Standards and Repositories

• Brazma, A, et al. Minimum information about a
  microarray experiment (MIAME)-toward standards
  for microarray data. Nature Genetics. 2001
  Dec29(4)373.
• http//www.nature.com/cgi-taf/DynaPage.taf?file/
  ng/journal/v29/n4/full/ng1201-365.html
• Ball, CA, et al. Submission of Microarray Data to
  Public Repositories. PLoS Biology. 2004
  September 2 (9) e317
• http//www.pubmedcentral.nih.gov/articlerender.fc
  gi?toolpubmedpubmedid15340489

33
Public GWAS Study Repositories

• dbGAP (database of genotype and phenotype) at
  the NLM
• http//www.ncbi.nlm.nih.gov/sites/entrez?dbgapcm
  dsearchterm

34
Central Dogma of Molecular Biology
DNA
RNA
Protein
Phenotype
Phenotype Tissues Organs Organisms
Genomics
Proteomics
Transcriptomics Functional Genetics
35
Proteome and Proteomics

• Proteome the entire set of proteins (and other
  gene products) made by the genome.
• Proteomics study of the interactions among
  proteins in the proteome, including networks of
  interacting proteins and metabolic
  considerations. Also includes differences in
  developmental stages, tissues and organs.

36
Protein Functions

• Catalysis
• Transport
• Nutrition and storage
• Contraction and mobility
• Structural elements
• Cytoskeleton
• Basement membranes

• Defense mechanisms
• Regulation
• Genetic
• Hormonal
• Buffering capacity

37
Protein Databases

• SwissProt
• PIR
  http//www.pir.uniprot.org/
• GENE http//www.ncbi.nlm.nih.gov/gene
• InterPro http//www.ebi.ac.uk/interpro/
• Correspond to (and derived from) Genome data
  bases
• All connected by Reference Sequences (NCBI)

UniProt
38
Gene/Protein Database entries

• HFE record in Entrez GENE (NCBI)
• http//www.ncbi.nlm.nih.gov/entrez/query.fcgi?db
  genecmdretrievedoptGraphicslist_uids3077

39
Structure Databases

• Contain experimentally determined and predicted
  structures of biological molecules
• Most structures determined by X-ray
  crystallography, NMR
• Example MMDB molecular modeling db
  http//www.ncbi.nlm.nih.gov/Structure/MMDB/mmdb.sh
  tml
• HFE Entry
• http//www.ncbi.nlm.nih.gov/Structure/mmdb/mmdbsrv
  .cgi?form6dbtDoptsuid9816

40
Protein Interaction Databases

• Record observations of protein-protein
  interactions in cells
• Attempts to detail interactions observed in
  thousands of small-scale experiments described in
  published articles
• Examples
• BIND Biomolecular Interaction Network Database
• DIP Database of Interacting Proteins
• MIPS Munich Information Center for Protein
  Sequences
• PRONET Protein interaction on the Web
• Many others, both academic and commercial

41
Controlled Vocabularies in Bioinformatics

• The Gene Ontology http//www.geneontology.org/
• Knowledge about gene function (the ontology
  itself)
• Annotation of gene products (for comparisons)
• The MGED Ontology (arising from MIAME)
• http//mged.sourceforge.net/
• Annotation of microarray experiments for public
  repositories
• Clinical Bioinformatics Ontology
• Annotation of gene tests in electronic medical
  records
• http//www.cerner.com/cbo
• MIAPE from Proteomics Standards Initiative (PSI)
• Annotation of proteomics experiments for public
  repositories
• http//psidev.sourceforge.net/

42
Genomics Data and Patient Care

• From genotype to phenotype

43
Human Disease Gene Specifics

• Genes linked to human diseases (9-2004)
• 425 in 2 yrs
• 1700/20,000 9 of loci

44
Informatics Issues related to Genomics Data and
Patient Care

• Linking known data for genes causing human
  diseases to clinical decision support and EMR
  documentation
• Linking treatment/dx data from gene expression
  profiles in EMRs
• Representation of genetic data in electronic
  medical records
• Making this understandable to providers and
  patients

45
Genetics is Impacting Medicine Today

• 2000 genes health conditions
• gt 1700 single gene tests for diagnosis
• http//www.Genetests.org
• Growing number of gene expression profiles
  (multi-gene microarray data)
• Relate to diagnosis, therapy, drug dosage,
  occupational hazards, reproductive plans, health
  risks, .

46
Well-known Examples (germ line)

• Pharmacogenetics
• CYP450 alleles exaggerated, diminished or
  ultra-rapid drug responses. E.G., Warfarin. 93
  of patients are OK on standard doses. 7 of
  patients have severe hemorrhage. CYP2C92 and
  CYP2C93 most severe of 6 known mutations.
• Environmental susceptibility
• Sickle Cell trait carrier and malaria parasite
• Nutrition
• PKU and avoidance of phenylalanine

47
(No Transcript)
48
Clinical Uses of Expression Profiling

• Disease Dx and Tx
• Distinguish morphologically similar cancers
• DLBCL (Poulsen et al (2005) Microarray-based
  classification of diffuse large B-cell lymphomas
  European Journal of Haematology 74(6)453-65.))
• Therapy potential
• Rabson AB, Weissmann D. From microarray to
  bedside targeting NF-kappaB for therapy of
  lymphomas. Clin Cancer Res. 2005 Jan 111(1)2-6.

49
More Applications

• Diagnostic tool to screen for infective agents
• Chip imprinted with set of pathogenic genomes
  used to identify bacterial, viral, or parasite
  genomic material in patients body fluids
• Diagnostic chip to check for mutations involved
  in drug-gene interactions.
• Roche Amplichip

50
Cardiac Transplant RejectionHoward Eisen MD,
Chief of Cardiology at Hahnemann (5-2009)

• Have 11 gene profile (the Allomap genes) to
  predict transplant rejection.
• http//www.xdx.com/allomap/
• Some centers reduce the numbers of biopsies if
  the molecular profile predicts low risk.
• New studies of other transplants (liver, kidney,
  etc)
• Brouard et al PNAS 2007 10415448-53. biomarker
  panel renal transplant rejection

51
Endomyocardial biopsy

• Currently Only way to test for rejection
• Risk Factors associated

52
Breast Cancer Gene Expression Profiling

• 4 main molecular classes of breast cancer using
  gene expression profiles (signatures)
• Basal-like
• HER2-positive
• Luminal B
• Luminal A
• MammaPrint commerically available genomic assay
  for outcome prediction
• http//en.wikipedia.org/wiki/MammaPrint
• Treatment based on profiles
• Sotiriou and Pusztai. NEJM 2009. 360790-800.
  Gene expression signatures in breast cancer.

53
Example (somatic mutation)Iressa (gefitinib)
erlotinib

• Non-small cell lung CA 140,000 pt/yr
• Iressa (Astra Zeneca) causes remission in 1 of 10
  patients. Newer drug is erlotinib
• Iressa erlotinib efficacy correlates with EGFR
  mutation in the tumor. Now have gene testing for
  EGFR so can target appropriate people.
  http//www.sciencemag.org/cgi/content/full/305/568
  8/1222a

54
Implications for Health Care System

• More gene tests will be ordered. reports of 300
  increase in gene tests in 2003.
• Arch Pathol Lab Med 2004, 128(12)1330-1333
• The FDA will regulate panels of tests.
• http//www.fda.gov/bbs/topics/news/2004/new01149.h
  tml
• Non-discrimination laws for insurance and
  employment will open a floodgate. GINA
• Preventive healthcare will play a larger part.
• Environmental risk factors dictate OSHA-type
  approach to worker empowerment and education
  about safe behavior

55
Unsolved Informatics IssuesWhat Should Be
Stored in the EMR?

• Complete DNA sequence for specific genes into the
  EMR? Where?
• Microarray expression and gene chip data?
• Meta-data about the DNA sequence arrays?
• If not the sequence (ie., diff from reference
  sequence), what to do when the reference sequence
  changes? Or gene chip changes?
• How to trigger alerts and reminders? And for
  what?
• How to link to the patients PHR?

56
Genetic data in electronic medical records

• Implications for component systems
• Laboratory
• Pharmacy
• Computerized order entry
• Documentation and notes
• Knowledge management
• Alerts and reminders
• Finding patients matching profiles
• Practice guidelines and clinical trials
• Appropriate therapies and medications
• Keeping current

57
Genome Data and Other Information Systems

• Genomic information will be pervasive in all
  healthcare information systems.
• Also in public health systems
• Newborn screening
• Tissue and organ banks
• DOD requires DNA samples
• Bioterrorism and homeland security
• Identification of World Trade Center victims
• Infection agent identification, origin spread
• Privacy and security issues are important but not
  inherently different than other EMR data.

58
Consumer Health Issues related to
Genetics/genomics
59
Genetics Home Reference

• Consumer health resource to help the public
  navigate from phenotype to genotype.
• Focus on health implications of the Human Genome
  Project.
• http//ghr.nlm.nih.gov
• Mitchell, Fun, McCray, JAMIA, 2004 Nov
  11(6)439-437

60
Direct to consumer genetic tests

• 23me https//www.23andme.com/
• Navigenetics http//www.navigenetics.com
• DecodeMe http//www.decodeme.com/

61
Reports on genotype-phenotype associations
62
(No Transcript)
63
What is Coming?

• Next generation sequencing
• More public information (personal genome project)
• Environmental variables to correlate with
  genotype functional info
• Human microbiome http//nihroadmap.nih.gov/hmp/
  
• Epigenetics
• Nanoparticles and nanomedicine
• More consumer activism
• Personalized Medicine
• And all of this in the EMR??

64
Summary

• Informatics will be the key enabling technology
  for personalized, genomic medicine.
• Current separation between bioinformatics and
  clinical informatics is diminishing as the two
  subdisciplines become more entwined

65
Discussion?
66
Optional ExerciseHands-on with GHR

• Scavenger hunt with hemochromatosis and the genes
  that influence it.
• Explore the Genetics Home Reference by answering
  the following questions. Start at
  http//ghr.nlm.nih.gov .

67
GHR Scavenger Hunt

• How common is hemochromatosis?
• How many genes have been proven to be involved in
  hemochromatosis when the genes are mutated?
• What are the symbols for these genes?
• Can you find the link to MedlinePlus with health
  information on hemochromatosis?

68
GHR Scavenger Hunt

• What are the names of the patient support
  associations for hemochromatosis?
• One synonym for this condition is bronze
  diabetes. Can you find a reason for this?
• What kind of damage is done to the liver of
  people with hemochromatosis?

69
GHR Scavenger Hunt

• For the genes involved in hemochromatosis, how
  many of them are available as a DNA test?
• Give one place where you would choose to send a
  tissue sample for DNA testing.
• What sites are listed under Research Resources
  for the TFR2 gene?
• How many alternately spliced proteins for TFR2?
• In what tissues is this gene expressed?

70
GHR Scavenger Hunt

• How do people inherit hemochromatosis?
• Do the genes involved in hemochromatosis cause
  other health conditions when they are mutated?
• Can you find a protein sequence for one of the
  genes?
• What clinical trials are available for
  hemochromatosis patients close to where you live?
  

71
Questions to

• Joyce Mitchell
• Joyce.mitchell_at_hsc.utah.edu
• http//uuhsc.utah.edu/medinfo